CFD Online Discussion Forums

CFD Online Discussion Forums (
-   Main CFD Forum (
-   -   outlet boundary condition (

Arash December 30, 2001 16:09

outlet boundary condition
In finite volume SIMPLE method which parameters must be defined for outlet boundary condition? thanks arash

John December 31, 2001 07:11

Re: outlet boundary condition
none unless you have multi-outlets - then you have to specify the proportion of mass flow out from each outlet.

Thomas P. Abraham January 1, 2002 22:42

Re: outlet boundary condition
Hello John,

Are you sure about that? I think, for an incompressible flow, we need to specify a constant pressure at the outlet.



John January 2, 2002 09:54

Re: outlet boundary condition
It really depends on what one means by outlet boundary condition: I understand it as a numerical boundary condition where flow and other variable (aside from pressure) changes very little in the streamwise direction, that is dfi/dx=0. Therefore one does not specify anything as the boundary condition will be determined as part of the solution. However if you mean outlet boundary condition as the flow goes out of the computational domain, then you can use many different specification such as pressure boundary.

STN January 4, 2002 00:17

Re: outlet boundary condition
It really depends on flow configuration. Is it steady of unsteady flow? For steady flow, normally, it use dPhi/dx = 0.0 which mean that there is no change along streamwise direction at the outlet.

Or for Unsteady flow, it use convective boundary condition, DPhi/Dt = dPhi/dt + Uc*dPhi/dx = 0.0 where Uc is convective velocity with mass conservation.

However, in the latter, I do not clear how to implement that equation into source code. Especialy for grid with zero volume on the boundary. Could anyone shed some light on me, please?

John January 4, 2002 07:02

Re: outlet boundary condition
The unsteady form of outlet convective boundary condition is majorly used when LES or DNS is used to prevent pressure wave reflected back into flow domain. Uc is choosen such that the overall mass is conserved and DPHI/Dx can be evaluated from (PHI_outlet-Phi_cell)/DX

Seb Perron January 4, 2002 09:47

Re: outlet boundary condition
Usually, you specify a free exit: the pressure is constant (there is no correction of the pressure at the outlet and it the pressure is usually equal to zero). Furthermore, the derivative of the streamwise velocity is zero ( neumann BC). For incompressible flow you most of the time follow this simple rule:

normal velocity imposed (inlet or wall), neumman BC for pressure correction;

normal velocity unknown (outlet) pressure must be given.

Hence, your problem will always be well posed.

All times are GMT -4. The time now is 17:45.